Skip to content

SynSem_PolymorphicVariadicPredicates

EmilyBender edited this page Apr 20, 2018 · 37 revisions

Problem Statement

In the ERG meaning representations (and in logic-based semantics, more generally), how to deal with variation in the number and types of semantic arguments?

Survey what other broad-coverage initiatives do, e.g. EngValLex,

Reading List

Resources to Review

  • ERG Lexicon

  • Prague valency lexicon

  • VerbNet

  • FrameNet

  • Levin, B. (1993) English Verb Classes and Alternations: A Preliminary Investigation, University of Chicago Press, Chicago, IL.

Clear Cases

  • Argument still there semantically even when dropped: eat

  • Argument gone semantically if not satisfied syntactically: kick

  • Argument can be NP or CP/S: report

  • Accepts clausal complement only: contend ... but FrameNet provides an example with and X contends the latter, so we need a better example here.

Examples

deny

FrameNet lists two frames for deny, Affirm_or_deny and Prevent_or_allow_possessionf. On examining the annotated data, we found the following valence frames, and aligned them with the disavow and decline like senses:

disavow (1,3) decline (2)
V-ing
S|CP
NP
NP NP
NP PP

With example arguments:

disavow (1,3) decline (2)
V-ing cheating
S|CP that they cheated
NP the rumor
NP NP the petitioners; the request; Kim a kiss
NP PP a kiss to Kim

Note1: In the NP, NP frame, either of the NPs can go missing, but not both at once: *Kim denied. Note2: FrameNet has one instance of something like deny a request classified with the frame that otherwise has the disavow senses. We think this is probably an annotation error.

Options

deny_v_1(e,x,h)                      deny_v_1(e,x,p)
deny_v_3(e,x,x)
deny_v_2(e,x,x,x)                    deny_v_2(e,x,x,x)

Would like to know how each of these options (as well as always getting the same predicate symbol) look to a logically-inclined semanticist. Which is preferable? Is there yet something different that would be better?

Observations: In this case at least, the analysis is consistent with the approach to having the Sem-I do the refining, since the information needed to distinguish between the predicate symbols is all apparent in the ERS. Exception: Since both ARG2 and ARG3 are syntactically optional in deny_v_2, how would we know which way to specialize NP denies NP? Alternatively, how do we know when to put in ARG3, especially given that in NP denies NP, the second NP could be either ARG2 or ARG3?

  • Kim denied the rumor.

  • Kim denied the request.

  • Kim denied the petitioners.

This entails two separate lexical entries, which could then have different PRED values.

More data:

  • Kim denied to Sandy that Pat arrived.

  • %Kim denied the rumor to Sandy. -- corpus study needed

  • *Kim denied to Sandy cheating.

  • *Kim denied cheating to Sandy.

Interesting that the addressee is available for CP complement but not the VP one. Necessarily unexpressed in VP frame, but still there in the semantics?

  • ?He denied my having received any bribes.

  • *He denied me having received any bribes.

Are gerunds similar in semantic type to CPs and nouns like rumor?

  • I resent that he sang that song.

  • I resent his singing of that song.

Is rumor something like a proposition in its semantics? But:

  • I deny most rumors that circulate about me. -- clearly quantified

  • Kim denied every annoying rumor. -- what is annoying modifying? Is it the content that's annoying? The frequency of the repetition?

  • I deny every false rumor. -- clearly the content is false.

  • The whispered rumors of his affairs

  • The frequent rumors of his affairs

Other Wiki Pages

Clone this wiki locally